Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Good system

  1. Aug 15, 2004 #1
    What are the characteristics of a good innovation/ system?
  2. jcsd
  3. Aug 16, 2004 #2

    Good question but a little hard to define. Some people thought that the Sony beta video tape system was great -- quality -- reliability -- resolution etc etc. and yet it was dumped -- why ? Sometimes it's summed up as
    could not record 5 hours of shlock TV !!!

    Seriously this question I believe may be divided in two
    1) professional
    2) entertainment -- I will leave that one alone

    a) It should WORK and do so reliably -- there is nothing worse than a system which you cannot rely on even if it's performance is a little less than you may hope -- a functioning 400 Mhz computer is a whole lot better than a 1 Ghz
    device which continually crashes , loses data , etc.
    However if there is a well defined specification to be met -- then that's what you must meet -- there's no point in going half way to the moon.

    b) Innovation yes , that is it should solve operational problems in smart highly useable ways ( i.e. good user interface where required ) some people and or companies manage to do this others not , it is a mentallity like saying the company President drives his own company cars.
    It's not always obvious an example is Corel Draw in which the 'object manager' is superb compared to Adobe and in complex drawing it's an absolute requirement -- it gives the impression that the Adobe people never use their own system.

    c) Innovation for it's own sake NO , sometimes to be 'on the cutting edge' people use unproven technology , which goes back to a)
    There is a lot of difference between research and development some mix the two due to lack of funding.

    d) Life cycle , appropriate to application , Sony has this correct .

    e) Cost / pricing
    Very important , technically good systems can also be too expensive , this requires excellent design concepts , innovative solutions , both of which are very people dependant both in detail AND in project management or overview.

    f) In conjunction with e) Marketing -- one must have a clear view of the market not just in the present tense but in the immerging sense , this requires detailed knowledge of what is going on within the relevant industry ( not guesswork or imagination ) and for some systems this means a very long term view ( say telephony or aircraft ).

    g) Good System -- good people it's a soft but vital part of the equation
    you need a mix of experience and innovative brash thinkers and someone to manage both.

    h) Good tools -- although you can go overboard on this especially where a
    a single tool may cost 100 K$ as in RF systems , a good workman makes sure his saw is sharp or at least has one available when needed.

    i) Styling , another soft but important issue , this is not a simple question of 'art' but a complex issue of functionallity as well as aesthetics , I remember one case where the fact that a system box was purposely designed
    for easy of entry to small commercial aircraft made a huge difference in shipping costs.

    j) Weight -- seems inocuous enough -- weight costs money in all sorts of ways and one thing in particular which adds weight is power consumption.
    This is also an area of innovation -- how do you remove weight by design
    even if the system will never be moved , it's a matter of materials , power supplies , heat dissipation and so on , shipping costs , and material storage .

    k) Repairability and testability
    Often the last thing on the engineers mind in development , but a major factor in practical use and production , ideally it should be a simple thing to verify system status and identify problems , obviously it helps if the system is partitioned into well defined sub systems which can be independantly dealt with -- tested replaced upgraded etc. but this is very application dependant
    in the consumer area they hardly care.

    These are just a few thoughts there are a plethora of others . Ray
    Last edited: Aug 16, 2004
  4. Aug 17, 2004 #3
    Good job, and How do you understand the word " Reliability " ? Do you know how much it influence the system?
  5. Aug 17, 2004 #4
    Reliability is a complex question -- in simple systems it is often ignored since modern components which are used in many different applications are themselves very good.
    However for complex systems this is a matter of statistical analysis using components which are characterised to mean time to failure under given conditions.
    Often in practice systems are 'burnt in' meaning they are stressed at temperature for a time to weed out 'weak' components.
    However the ultimate meaning has to do with application ( and therefore design)
    a nuclear power control circuit is a little different from your average PC and is treated differently.
    This is a serious subject dealt with by specialists.
  6. Aug 19, 2004 #5
    Let's not forget about 5S and ISO!

    Very important parts good system. You have to have organization and documentation and following the guidelines laid out in these two systems will help to achieve that.
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?

Similar Discussions: Good system